16

Whenever I see "planes are safer than driving by car", it's always based under the metric of by "miles travelled per crash", or "journeys per crash" based. The latter might be bad because it may not account for that we make far fewer trips on a plane than a car, and also may discount that multiple people may die if the plane crashes. Even if we put those in, people have different transport requirements and may drive or fly near or far. It's a crude metric overall.

"Miles travelled per crash" may also be misleading due to the multiple people on board a plane, and even if you use per capita (person), I'd argue that many people intuitively prefer to go by time instead of miles when travelling, at least on some occasions. We tend to ignore how far the destination is, and instead just look at how long we're spending on the mode of transport.

So to reiterate the title, is there a list of the least to most dangerous forms of transport where we take into account the time spent upon that vehicle, per person?

17
  • 7
    I think that it is usually deaths per passenger mile. So, a plane crash won't be equivalent to a single car crash but hundreds of car crashes.
    – badjohn
    Commented Oct 26, 2023 at 22:42
  • 2
    Also, the difference between plane and train safety is so great that planes will come out better in either measure. Consider trains which are also much safer than cars.
    – badjohn
    Commented Oct 26, 2023 at 22:44
  • 2
    In the previous comment, it should have been "between plane and CAR safety".
    – badjohn
    Commented Oct 26, 2023 at 23:47
  • 13
    Note that if you want to chose the safest option between different modes of transport you generally pick between options that bring you from some point A to some point B. To meaningfully compare them you should look at deaths per passenger mile not at deaths per hour in transport. There is a good reason deaths per passenger mile is the standard for measuring this, it is the scale that allows you to make useful comparisons.
    – quarague
    Commented Oct 27, 2023 at 9:44
  • 8
    Remembering that take off and landing are by far the most dangerous parts of air travel, I don't see how comparing a 14 hour flight (with one take off/landing) to 14 hours of driving (where each passing moment is getting more dangerous as I'm growing more tired toward the end) is in any way, shape, or form a valid comparison.
    – FreeMan
    Commented Oct 27, 2023 at 13:38

2 Answers 2

11

I have found another source for risk by hour of travel, from Turbli.

The Safest Transport Modes, Ranked by Statistics From 10 Years of Data.

The analysis is done for the US only, using the data provided by the Bureau of Transportation Statistics (BTS).

It includes risk by absolute fatalities, by passenger mile, and –

Fatalities by passenger hours

enter image description here

The chart isn't sorted by risk, but you can see it is broadly similar to the data I showed in another answer. However the chart is visually distorted because the length of each bar isn't proportional to its statistic.

The linked article has a lengthy discussion about travel risk. It mentions these sources:

US Bureau of Transportation Statistics. National Transportation Statistics.
US Federal Transit Administration. Annual National Transit Summaries and Trends.
Savage I., 2013. Comparing the fatality risks in United States transportation across modes and over time. Research in Transportation Economics, 43, 9-22.


It seems that my earlier answer used a UK data set and this one uses a US data set.
It can be seen that bus, rail, car and motorcycle are in the same sequence of risk.

But a significant difference is the US data making 3 categories for air travel.

  • "scheduled flight with more than 10 passengers" is safer than by bus.
  • "scheduled flight with less than 10 passengers" is worse than by car.
  • "private plane" is much worse, more than half the risk of a motorbike.

The UK data presumably averages all three categories, but "scheduled flight with more than 10 passengers" is what most people travelling will think of by "flying".

18
  • 7
    Looks like the chart is in log scale. I guess its author didn't find it necessary to add any tick marks to make this clear.
    – Anyon
    Commented Oct 28, 2023 at 23:34
  • 8
    @Anyon Just for interests sake, this is the graph reproduced to show log vs linear scale. You'll have to forgive the log scale showing ticks of 5's rather than 10s, quirk of Excel. Commented Oct 29, 2023 at 13:18
  • 1
    bus, rail, car and motorcycle are in the same sequence of risk => seems like bus and rail are 10x safer than car and car is 50x safer than motorcycle?
    – JonathanReez
    Commented Oct 29, 2023 at 14:48
  • 1
    Because the smaller planes are more likely to crash, and the private planes even more so (per passenger-hour)? You asked if there are any statistics, and I have provided two independent sets. I didn't compile them, so my guesses are as good as yours. Commented Oct 29, 2023 at 23:20
  • 1
    @DanW There's a minor typo in the chart, footnote 4 should read in parenthesis 14 CFR 121 and 5 should read 14 CFR 135. They're actually very specific, "Part 121" refers to the very strict rules airlines run under, and "Part 135" refers to other commercial air carriers. Major differences include a special pilot's license for Part 121 that requires more flying time, and Part 121 requires at least 2 engines, where Part 135 planes can have one.
    – user71659
    Commented Oct 30, 2023 at 2:59
30

This page from REStARTS has a table (emphasis original):

7. The risk of travel

There are three possible ways of quoting transport risk; in terms of distance travelled, number of journeys or time of travel. Interested parties tend to choose the form of presentation that suits their own purposes.

The air transport industry, for example, will almost always choose a per km basis, which is optimum for them, as most fatalities occur on landing and take-off, while the distances are large. Land based transport organisations, in contrast, will tend to select fatalities per number of journeys or hours of travel, since the risks are uniformly spread. Thus both are able to demonstrate that theirs is the safest form of transport. The actual statistics are given below (taken from an article by Roger Ford in Modern Railways, Oct 2000 and based on a DETR survey). They record the number of fatalities per billion km, billion journeys or billion hours of travel.

km  journeys  hours
Air 0.05  Bus 4.3  Bus 11.1
Bus 0.4  Rail 20  Rail 30
Rail 0.6  Van 20  Air 30.8
Van 1.2  Car 40  Water 50
Water 2.6  Foot 40  Van 60
Car 3.1  Water 90  Car 130
Pedal cycle 44.6  Air 117  Foot 220
Foot 54.2  Pedal cycle 170  Pedal cycle 550
Motorcycle 108.9  Motorcycle 1,640 Motorcycle 4,840

So if you are concerned about risk: bus is safest.

In the first column (by km) air is safest, and significantly less safe in the other columns. Even though bus isn't a viable alternative for those long distance flights it is still second. Apart from that, bus is safer than all other modes of transport, whatever way the stats are presented, and rail comes next.


Regarding the original sources, I looked through the many pages of articles by Roger Ford in Modern Railways, and there were 3 in October 2022, the most likely being Informed Sources: Constraints needed for accident investigation closures but the site needs a subscription to read the article. I also searched on Department of the Environment, Transport and the Regions (DETR) at The National Archives but found nothing for "travel safety air bus train".

11
  • Comments have been moved to chat; please do not continue the discussion here. Before posting a comment below this one, please review the purposes of comments. Comments that do not request clarification or suggest improvements usually belong as an answer, on Travel Meta, or in Travel Chat. Comments continuing discussion may be removed.
    – Willeke
    Commented Oct 27, 2023 at 19:33
  • Oh! Sorry, the article by Roger Ford was in 2000, but the Modern Railways online archive only seems to go back to 2020. Commented Oct 28, 2023 at 19:11
  • (Chat room is closed and nobody there made this argument anyway.) I am not sure how you can arrive at the summary "So if you are concerned about risk: bus is safest.". Very clearly, if you are concerned about risk you should fly because it is about an order of magnitude safer. I say this so confidently because fatalities/journey and fatalities/hour compare apples with pears (average trip size is very different, and speed is very different). If you are facing a specific trip the only thing that counts is the risk for that trip which is solely determined by distance. Commented Dec 23, 2023 at 12:29
  • As an example, if you plan a trip around the world, walking (which would take years) would be much much riskier than flying which would take about two or three days. That is true even if risk per journey is not very good for planes. Well, yes, because those trips are much longer. Fore a specific risk, you must look at the specific trip. Yes, if you fly between continents, your risk is higher than when you walk to the grocer. That is what is reflected in the journey column but it is not what should inform your decision concerning a specific trip. Commented Dec 23, 2023 at 12:33
  • @Peter-ReinstateMonica the 'by km' column shows that air is safest from that point of view, but you won't be doing long journeys by bus anyway. Bus is at the top of the 'per journey' and 'per hour' columns. That's why I said it was safest. You've cherry-picked a quite unusual journey. Commented Dec 23, 2023 at 12:33

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .